Research map / Label-efficient and robust learning / Recognition and long-tail
Generative View-Correlation Adaptation for Semi-Supervised Multi-View Learning
Yunyu Liu, Lichen Wang, Yue Bai, Can Qin, Zhengming Ding, Yun Fu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add SeMix (semi-supervised mixup) instead of Mixup (ablation) | MLP · DHA RGB+Depth | accuracy (%): 81.48 -> 83.15 | Improved |
| add SeMix to baseline MLP (ablation) | MLP · DHA RGB+Depth | accuracy (%): 79.12 -> 83.15 | Improved |
| add cross-view adaptation (VCA) (ablation) | MLP · DHA RGB view | accuracy (%): 77.10 -> 75.26 | Hurt |
| add cross-view adaptation (VCA) (ablation) | MLP · DHA RGB+Depth | accuracy (%): 79.12 -> 81.32 | Improved |
| use entropy-based confidence weighting in VCA (VCA-entropy) (ablation) | VCA · DHA RGB+Depth | accuracy (%): 81.32 -> 84.10 | Improved |
| add SeMix and VCA-entropy together (complete model) (ablation) | VCA-entropy · DHA RGB+Depth | accuracy (%): 84.10 -> 89.31 | Improved |
| add generated samples at 1x ratio (ablation) | Ours · DHA | accuracy (%): 84.10 -> 89.31 | Improved |
| add generated samples at 1x ratio (ablation) | Ours · UWA | accuracy (%): 75.49 -> 77.08 | Improved |
| add generated samples at 1x ratio (ablation) | Ours · MHAD | accuracy (%): 98.23 -> 98.94 | No clear effect |
| increase generated samples from 2x to 3x (ablation) | Ours · UWA | accuracy (%): 77.47 -> 76.28 | Hurt |
| use limited labeled data (10%) with Ours vs GMVAR | Ours · DHA (10% labels) | accuracy (%): 44.86 -> 48.15 | Improved |
| compare Ours with GMVAR with full labels | Ours · MHAD | accuracy (%): 98.94 -> 98.94 | No clear effect |
Closest papers by meaning
- Semi-Supervised Multi-View Multi-Label Learning with View-Specific Transformer and Enhanced Pseudo-Label (AAAI 2025 · similarity 0.89)
- Multi-View Multi-Label Classification via View-Label Matching Selection (AAAI 2025 · similarity 0.85)
- CFDM: Contrastive Fusion and Disambiguation for Multi-View Partial-Label Learning (AAAI 2025 · similarity 0.85)
- Beyond Random Augmentations: Pretraining with Hard Views (ICLR 2025 · similarity 0.84)
- Incomplete Multi-View Multi-label Learning via Disentangled Representation and Label Semantic Embedding (CVPR 2025 · similarity 0.84)
- Partial Multi-View Multi-Label Classification via Semantic Invariance Learning and Prototype Modeling (ICML 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
